PSYCH 210 Lecture 15 Outline of Last Lecture I. Hypothesis testing and z-testsa. Finish factors affecting powerb. Power calculation for xII. One-sample t-testOutline of Current Lecture I. One-sample t testa. Variation: no μb. AssumptionsII. Independent-sample t testa. Rationale for new testb. New formula c. Dog Training ExamplesCurrent LectureI. One-sample t testa. Continue Examplei. Hypothesesii. Critical Values of tiii. Calculate Observed tiv. Make decision regarding H0These notes represent a detailed interpretation of the professor’s lecture. GradeBuddy is best used as a supplement to your own notes, not as a substitute.1. Reject H0a. 3.15 > 2.262v. State Conclusion1. Tx/IV/levels2. Specific DV3. Sig difference?4. Full Comparisona. Ex) The presence of El Nino resulted in a significant difference between the mean bill length in sample and original population mean bill lengthb. Variation on one-sample ti. μ is not given, but must be calculated based on what is expected by change1. Ex) Eyespots in prey species (e.g. false-eyed frog)a. Question: Determine whether the presence of eyespots influences the amount of time birds spend in a test chamberi. DV=time bird spends on left side of chamberii. Test for 60 miniii. Therefore, μ = 30 minb. Hypothesesi. H0=30ii. H1≠30c. Assumptionsi. Interval or ratio dataii. Independent observationsiii. Normal distributionII. Independent-sample t testa. Rationale for new test (Why do we need a new test?)i. For z and one-sample t1. Compare a single sample mean (M) to μ2. Only one treatment; compare to untreated population3. Need info on population parameters a. z: μ, σb. t: μii. We often want to use TWO treatment groups, not just one1. Compare two sample groups to each otheriii. We often don’t have any info about population parametersb. Criteria for Independent-Sample t-testi. No population informationii. Two Tx groups (2 M’s)1. M1 – M2iii. Different Participants in each groupc. Theoryi. Testing whether 2 sample M’s come from the same or different populations1. Are they so different that they had to have come from separate populations? (=Reject Null)2. Or are they so similar that they share the same population? (=Fail to Reject)d. New Formulai. One-sample t=(M-μ)/sMii. Independent-samples t = [(M1-M2)-(μ1-μ2)]/SM(1-2)1. Set (μ1 – μ2) to the expected difference under H0 = 02. Pooled Variance e. Examples from dog trainingi. Methods for teaching ‘Leave it’1. Leash correction (LC) vs. body blocking (BB)a. Tx1=LC, Tx2=BBb. n1 =20, n2 =20c. DV:i. Test: Put item down, say cue1. # of s left item aloneii. Test: # of successful trials/10ii. Call command1. Tone of Voice (IV)a. “Stern” vs. “Happy” (Levels)2. Movement a. None vs. Moving Away 3. Use of Rewarda. No treat vs. Treat4. Body Posturea. Upright/Forward vs. Low/Awayiii. Train ‘Stay’1. Timing of Rewarda. Treat at end vs. Treat while lying
View Full Document
Unlocking...